A major bottleneck for studying biomolecular systems is the accuracy and availability of consistent molecular mechanical (MM) models, which in turn depend on accurate representation of the potential energies (referred to as the “force field (FF)”). A high-quality force field is the key to successfully model the structures, energies and dynamics of biomolecules as well as to successfully describe biological processes on many levels (folding, binding, cell signaling, etc.) using molecular dynamics (MD) simulations.

AMBER (AMBER Homepage) is a main stream force field in molecular dynamics simulations of biomolecules. We have devoted ourselves in AMBER force field development for many years. Our force fields (GAFF, Parm99, FF12pol, etc.) have been extensively used by our users all over the world. The effective functional form, shown in below, is very robust and appliedby many AMBER force fields (FF94/FF99 series, GAFF, etc).

The following is an example of AMBER force field for the phenylalanine residue: 9 atom types are defined, partial atomic charges (red digits) are derived to reproduce electrostatic potential calculated at HF/6-31G* level; in total, there are 10, 15 and 21 parameters for the bonds, angles and torsional angles, respectively.

General AMBER Force Field (GAFF). Computer-aided drug design is an indispensable technique in modern drug discovery. We developed the original version of GAFF to meet the need to permit the development of high-quality general purpose force field parameters to study protein-ligand interactions. GAFF has proven to be a very successful force field as demonstrated by its extensive number of citations (more than 2,900 times according to Web of Science).

We have started and will continue to develop the second generation of the general AMBER force field - GAFF2. Specifically, We plan to (i) redevelop the van der Waals parameters to reproduce both the high-quality interaction energies and key liquid properties such as density, the heat of vaporization, and hydration free energy; (ii) extend the coverage of chemical space to all the nonmetal elements (except noble gases) and common metals; (iii) parameterize the bonded force field parameters (bond stretching, bond angle bending and torsional twisting) using high-quality ab initio data; and (iv) improve the transfer-ability of GAFF2 by including many more model molecules for force field parameterizations. We strongly believe that GAFF2 will be an even more successful general purpose force field and that GAFF2-based scoring functions will significantly improve the success rate of virtual screenings.

The parameterization strategy is summarized in MMFF . An example of deriving torsional angle parameters for the O=C-O-H using acetic acid as the model compound is demonstrated below.

PTM Force Fields (PTMFF). Post-translational modifications (PTMs) are chemical modifications of proteins that are critically important in regulating a protein’s structure and function.10 There are more than 400 types of PTMs that affect more than 70% of proteins. However, the widely-used software packages for biomolecular simulations only have force field parameters for the standard and a limited number of non-standard amino acid residues. We are now developing a PTM force field (PTMFF) that covers the common PTMs (phosphorylation, acetylation, methylation, glycosylation, etc.) and is totally compatible with the existing biomolecular force fields. The PTMFF will be validated by studying the structures, functions, and dynamics of PTM decorated proteins through MD simulations. We are developing PTMFF following the canonical AMBER force field development strategy. A web toolkit deploying PTMFF is accessible by click DPTMFF Database

AMBER Polarizable Force Field. Explicit inclusion of polarization effects (in contrast to a fixed charge on each atom) in a molecular mechanical energy function is critical to model heterogeneous and highly charged systems, such as membrane proteins, nucleic acids, etc. With the support of the NIH R01 grant "AMBER force field consortium", we are now developing the polarizable general AMBER force field (pGAFF) targeted to accurately calculate the free energies of protein-ligand complexes. To facilitate the use of pGAFF in virtual drug screening, a fast method that can automatically generate partial charges for any arbitrary organic molecule is also being developed. Besides the pGAFF development, we also plan to investigate the potential energy terms for charge transfer, electron penetration, and quadrupole moments in order to lay a solid foundation for the development of a force field that will take full account of these contributions.

Antechamber – A Molecular Mechanical Toolkit. To enable accurate MM-based studies, especially in rational drug design, it is critical to develop a set of reliable tools so that force field parameters can be generated for arbitrary molecules accurately and automatically. 

I developed Antechamber to fulfill this mission. The Antechamber module has been released with AMBER since Version 8. Since then the Antechamber module has been enjoying broad applications both among and outside the AMBER community (there are more than 750 SCI citations, however, the number of publications using Antechamber is much higher as most papers only cited the AMBER package instead of the Antechamber module itself). In order to broaden the application of Antechamber and to appropriately cope with the extended chemical space in GAFF2, we are now redeveloping some key Antechamber algorithms. Additional functions such as the automatic generation of non-standard residue topologies will be introduced. 

Web toolkit for MM Calculations. To broaden the application of GAFF and Antechamber, I have developed a web-based toolkit (using MySQL/PHP) to enable users to generate high quality MMFF parameters for arbitrary molecules. I are developing a molecular mechanical model (MMM) database to store MM models for drugs, actives, co-factors and drug-like molecules.